Inspiration
We find it quite challenging to look for new foods nearby; there's just too many options to explore! That's why we created this web app that filters the selection of eateries just based on a food picture.
What it does
The user has to upload an image of a food/foods they want to eat. Our app will analyse it and filter restaurants based on the image.
How we built it
We used YOLOv8 to identify foods in images, and integrated it into a web app using Flask. Additionally, the restaurant data was web scraped using beautiful soup. We used leaflet.js and OSM to integrate a map into the web app to view the location of nearby eateries.
Challenges we ran into
There was too little time for how much we wanted to accomplish. Furthermore, we had little to no experience with the fields of computer vision and web scraping. All these new concepts were very challenging to learn.
Accomplishments that we're proud of
We managed to learn a lot of new concepts in just under 24 hours! Our web app has been developed by just us two team members who did not know almost all these concepts before the project.
What we learned
From management planning to learning computer vision and web scraping, we managed to accomplish a lot of new achievements this Durhack 2024. The beginning was overwhelming, with the amount of people and information loaded onto us. We learned how to relax and take this hackathon as more of a fun challenge!
What's next for NottsFoodFinder
We want to be able to filter eateries more accurately in the future. NottsFoodFinder will allow the user to view nearby eateries based on their location, and filter them based on reviews. Furthermore, our AI model will be improved with a larger dataset and vigorous training. More information about different restaurants will be included. This will make NottsFoodFinder an exciting experience for foodies!
Log in or sign up for Devpost to join the conversation.